Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Cloudius Systems presents: Seastar Avi Kivity, April 13 2015
● New tech, runs on physical machines, VMs,Linux/OSv ● Multi-million IOPS, fully scalable ● Perfect building block for dat...
SeaStar current performance
SeaStar Before: Thread model After: SeaStar shards
Problem with today’s programing model + Single core performance (frequency, IPC) no longer growing + #core grows but it’s ...
SeaStar Framework Linear scaling by #core + Each engine is executed by each core + Shared-nothing per-core design + Fits e...
Kernel SeaStar Framework Comparison Application TCP/IPScheduler queuequeuequeuequeuequeue threads NIC Queues Kernel Tradit...
SeaStar handles 1,000,000s connections in parallel! Traditional stack SeaStar’s sharded stack Promise Task Promise Task Pr...
SeaStar current performance Stock TCP stack SeaStar’s native TCP stack
Basic model ■ Futures ■ Promises ■ Continuations
F-P-C defined: Future A future is a result of a computation that may not be available yet. ■ Data buffer from the network ...
F-P-C defined: Promise A promise is an object or function that provides you with a future, with the expectation that it wi...
Basic future/promise future<int> get(); // promises an int will be produced eventually future<> put(int) // promises to st...
Chaining future<int> get(); // promises an int will be produced eventually future<> put(int) // promises to store an int v...
Zero copy friendly future<temporary_buffer> connected_socket::read(size_t n); ■ temporary_buffer points at driver-provided...
Zero copy friendly (2) future<size_t> connected_socket::write(temporary_buffer); ■ Future becomes ready when TCP window al...
Dual Networking Stack Networking API Seastar (native) Stack POSIX (hosted) stack Linux kernel (sockets) User-space TCP/IP ...
Disk I/O ■ Zero copy using Linux AIO and O_DIRECT ■ Some operations using worker threads (open() etc.) ■ Plans for direct ...
Rich APIs ● HTTP Server ● HTTP Client ● RPC client/server ● map_reduce ● parallel_for_each ● distributed<> ● when_all() ● ...
More info ■ http://github.com/cloudius-systems/seastar ■ http://seastar-project.com
Thank you @CloudiusSystems
Upcoming SlideShare
Loading in …5
×

Back to the future with C++ and Seastar

23,310 views

Published on

Seastar is a modern, open source server application framework written in C++ that presents a future/promise based API to the user while delivering top-of-the line performance -- more than five times the nearest competitor, with 7 million requests per second served on a single machine.

Published in: Software
no profile picture user

  • Be the first to comment

Show More

Back to the future with C++ and Seastar

  1. 1. Cloudius Systems presents: Seastar Avi Kivity, April 13 2015
  2. 2. ● New tech, runs on physical machines, VMs,Linux/OSv ● Multi-million IOPS, fully scalable ● Perfect building block for database/filesystem/cache ● Share-nothing, fully asynchronous model ● Open Source SeaStar Technology
  3. 3. SeaStar current performance
  4. 4. SeaStar Before: Thread model After: SeaStar shards
  5. 5. Problem with today’s programing model + Single core performance (frequency, IPC) no longer growing + #core grows but it’s hard to utilize. Apps don’t scale + Locks have costs even w/o contention + Data is allocated on one core, copied and used on others + Software can’t keep up with the recent hardware (SSD, line rate for 10Gbps, NUMA, etc) Kernel Application TCP/IPScheduler queuequeuequeuequeuequeue threads NIC Queues Kernel Traditional stack Memory
  6. 6. SeaStar Framework Linear scaling by #core + Each engine is executed by each core + Shared-nothing per-core design + Fits existing shared-nothing distributed applications model + Full kernel bypass, supports zero-copy + No threads, no context switch and no locks + Instead, asynchronous lambda invocation Application TCP/IP Task Scheduler queuequeuequeuequeuequeuesmp queue NIC Queue DPDK Kernel (isn’t involved) Userspace Application TCP/IP Task Scheduler queuequeuequeuequeuequeuesmp queue NIC Queue DPDK Kernel (isn’t involved) Userspace Application TCP/IP Task Scheduler queuequeuequeuequeuequeuesmp queue NIC Queue DPDK Kernel (isn’t involved) Userspace Application TCP/IP Task Scheduler queuequeuequeuequeuequeuesmp queue NIC Queue DPDK Kernel (isn’t involved) Userspace
  7. 7. Kernel SeaStar Framework Comparison Application TCP/IPScheduler queuequeuequeuequeuequeue threads NIC Queues Kernel Traditional stack SeaStar’s sharded stack Memory Lock contention Cache contention NUMA unfriendly Application TCP/IP Task Scheduler queuequeuequeuequeuequeuesmp queue NIC Queue DPDK Kernel (isn’t involved) Userspace Application TCP/IP Task Scheduler queuequeuequeuequeuequeuesmp queue NIC Queue DPDK Kernel (isn’t involved) Userspace Application TCP/IP Task Scheduler queuequeuequeuequeuequeuesmp queue NIC Queue DPDK Kernel (isn’t involved) Userspace Application TCP/IP Task Scheduler queuequeuequeuequeuequeuesmp queue NIC Queue DPDK Kernel (isn’t involved) Userspace No contention Linear scaling NUMA friendly
  8. 8. SeaStar handles 1,000,000s connections in parallel! Traditional stack SeaStar’s sharded stack Promise Task Promise Task Promise Task Promise Task CPU Promise Task Promise Task Promise Task Promise Task CPU Promise Task Promise Task Promise Task Promise Task CPU Promise Task Promise Task Promise Task Promise Task CPU Promise Task Promise Task Promise Task Promise Task CPU Promise is a pointer to eventually computed value Task is a pointer to a lambda function Scheduler CPU Scheduler CPU Scheduler CPU Scheduler CPU Scheduler CPU Thread Stack Thread Stack Thread Stack Thread Stack Thread Stack Thread Stack Thread Stack Thread Stack Thread is a function pointer Stack is a byte array from 64k to megabytes Context switch cost is high. Large stacks pollutes the caches No sharing, millions of parallel events
  9. 9. SeaStar current performance Stock TCP stack SeaStar’s native TCP stack
  10. 10. Basic model ■ Futures ■ Promises ■ Continuations
  11. 11. F-P-C defined: Future A future is a result of a computation that may not be available yet. ■ Data buffer from the network ■ Timer expiration ■ Completion of a disk write ■ Result computation that requires the values from one or more other futures.
  12. 12. F-P-C defined: Promise A promise is an object or function that provides you with a future, with the expectation that it will fulfil the future.
  13. 13. Basic future/promise future<int> get(); // promises an int will be produced eventually future<> put(int) // promises to store an int void f() { get().then([] (int value) { put(value + 1).then([] { std::cout << "value stored successfullyn"; }); }); }
  14. 14. Chaining future<int> get(); // promises an int will be produced eventually future<> put(int) // promises to store an int void f() { get().then([] (int value) { return put(value + 1); }).then([] { std::cout << "value stored successfullyn"; }); }
  15. 15. Zero copy friendly future<temporary_buffer> connected_socket::read(size_t n); ■ temporary_buffer points at driver-provided pages if possible ■ discarded after use
  16. 16. Zero copy friendly (2) future<size_t> connected_socket::write(temporary_buffer); ■ Future becomes ready when TCP window allows sending more data (usually immediately) ■ temporary_buffer discarded after data is ACKed ■ can call delete[] or decrement a reference count
  17. 17. Dual Networking Stack Networking API Seastar (native) Stack POSIX (hosted) stack Linux kernel (sockets) User-space TCP/IP Interface layer DPDK Virtio Xen igb ixgb
  18. 18. Disk I/O ■ Zero copy using Linux AIO and O_DIRECT ■ Some operations using worker threads (open() etc.) ■ Plans for direct NVMe support
  19. 19. Rich APIs ● HTTP Server ● HTTP Client ● RPC client/server ● map_reduce ● parallel_for_each ● distributed<> ● when_all() ● timers
  20. 20. More info ■ http://github.com/cloudius-systems/seastar ■ http://seastar-project.com
  21. 21. Thank you @CloudiusSystems

×